Pharmacology, Biochemistry and Behavior 124 (2014) 425–433

Contents lists available at ScienceDirect

Pharmacology, Biochemistry and Behavior

journal homepage: www.elsevier.com/locate/pharmbiochembeh

Fluoxetine and WAY 100,635 dissociate increases in scototaxis and

analgesia induced by conspecific alarm substance in zebrafish
(Danio rerio Hamilton 1822)
Caio Maximino a,b,⁎, Monica Gomes Lima a,b, Carina Cardoso Costa c,
Iêda Maria Louzada Guedes c,d, Anderson Manoel Herculano b,e
a
Departamento de Morfologia e Ciências Fisiológicas, Universidade do Estado do Pará, Marabá, PA, Brazil
b
International Zebrafish Neuroscience Research Consortium, USA
c
Programa de Educação Tutorial PET Medicina/Enfermagem, Instituto de Ciências da Saúde, Universidade Federal do Pará, Belém, PA, Brazil
d
Laboratório de Biologia Celular e Molecular do Citoesqueleto, Instituto de Ciências Biológicas, Universidade Federal do Pará, Belém, PA, Brazil
e
Laboratório de Neuroendocrinologia, Instituto de Ciências Biológicas, Universidade Federal do Pará, Belém, PA, Brazil

a r t i c l e i n f o a b s t r a c t

Article history: Alarm reactions to a substance secreted by the damaged skin of conspecifics and closely-related species are in-
Received 17 September 2013 creasingly being recognized as fear-like responses in fish. The neurochemical underpinnings of these effects
Received in revised form 30 June 2014 are so far unknown; however, given the role of the serotonergic system on defensive behavior, it is possible
Accepted 6 July 2014
that the alarm reaction is mediated by this monoamine. Exposure to conspecific alarm substance (CAS) increased
Available online 11 July 2014
anxiety-like behavior in the light/dark test in zebrafish and decreased nocifensive behavior. These effects were
Keywords:
accompanied by increases in blood glucose, hemoglobin, epinephrine and norepinephrine levels, as well as extra-
Alarm substance cellular levels of serotonin in the brain. Pretreatment with fluoxetine blocked the anxiogenic effects of CAS on the
Fear light/dark test as well as all physiological parameters and the increase in extracellular brain 5-HT, but not the
Scototaxis reduction in nocifensive behavior. Conversely, pretreatment with the 5-HT1AR antagonist WAY 100635 blocked
Nocifensive behavior the effects on nocifensive behavior, but not the effects on anxiety-like behavior nor on physiological parameters.
Zebrafish These results point to an important and complex role of the serotonergic system in the mediation of fear-
Serotonin potentiated behavior in the light/dark test and in fear-induced analgesia in zebrafish.
© 2014 Elsevier Inc. All rights reserved.

1. Introduction
A dual role for serotonin (5-HT) has been proposed in the control of
mammalian defensive behavior, with the neurotransmitter increasing
anxiety-like and decreasing fear-like behavior (Graeff et al., 1996,
1997). Although it is yet unknown if such a behavioral specialization
exists in teleost fish (Kalueff et al., 2012), two of the most widely used
behavioral tasks in zebrafish – the novel tank test and the light/dark
test – show differential pharmacological sensitivity to anxiolytic and
panicolytic drugs: dark preference is sensitive to anxiolytic, but not
panicolytic drugs (Maximino et al., 2011), while bottom-dwelling is
also sensitive to panicolytic drugs (Stewart et al., 2011). Besides their
differential pharmacological sensitivity, both tests are also under differ-
ent stimulus control; while the light/dark preference test is controlled
by an approach/avoidance conflict (Blaser et al., 2010; Maximino et al.,
2010; Blaser and Peñalosa, 2011; Araujo et al., 2012), the novel
⁎ Corresponding author at: Departamento de Morfologia e Ciências Fisiológicas,
Universidade do Estado do Pará, Campus Marabá Av. Hiléia Agrópolis do INCRA, s/n,
68503-120 Marabá, PA, Brazil.
E-mail address: caio.maximino@gmail.com (C. Maximino).
tank test is controlled by escape from the surface (Blaser and
Goldsteinholm, 2012; Luca and Gerlai, 2012). These observations closely
parallel Gray's model in which anxiety involves defensive approach,
generating an approach–avoidance conflict, while fear involves defen-
sive avoidance, generating escape/withdrawal responses (Gray and
McNaughton, 2000; McNaughton and Corr, 2004). Interestingly, in-
creasing 5-HTergic transmission by blocking uptake has been shown
to reduce bottom-dwelling but increase white avoidance in adult
zebrafish (Sackerman et al., 2010; Maximino et al., 2013a), while 5-HT
depletion produces the opposite effect (Maximino et al., 2013a).
These behavioral observations suggest different stimulus control,
and coordinate well with the “dual role” of serotonin proposed by
Deakin and Graeff (1991), Graeff et al. (1997), and Guimarães et al.
(2010). However, the differences observed in the effects of fluoxetine
and pCPA in both tests (Maximino et al., 2013a) could also be due to
other factors, including, e.g., alterations in contrast discrimination
(in the case of dark preference) or effects on the swim bladder (in the
case of bottom-dwelling). To better discriminate these hypotheses, a
clearer fear-inducing stimulus is needed. A behavioral response to
an “alarm substance” released by the damaged skin of conspecifics
(conspecific alarm substance, CAS), dubbed Schrecksreaktion by Von

http://dx.doi.org/10.1016/j.pbb.2014.07.003
0091-3057/© 2014 Elsevier Inc. All rights reserved.
426 C. Maximino et al. / Pharmacology, Biochemistry and Behavior 124 (2014) 425–433
Frisch (1938), has been proposed in zebrafish as a model to study innate
fear (Jesuthasan and Mathuru, 2008; Gerlai, 2010). In addition to in-
creased bottom-dwelling (Egan et al., 2009), shoal cohesion (Speedie
and Gerlai, 2008) and erratic movements in the novel tank test and
group behavior task (Speedie and Gerlai, 2008; Egan et al., 2009;
Mathuru et al., 2012), CAS has been demonstrated to promote analgesia
in zebrafish (Maximino, 2011; Lima et al., 2012) as well as in the piauçu
Leporinus macrocephalus (Alves et al., 2013). In this latter species,
acute treatment with fluoxetine (10 mg/kg) reverts the locomotory-
inhibiting effect of exposure to CAS (Barbosa et al., 2012). In the crucian
carp, Schrecksreaktion is in part mediated by structures which receive
primary and secondary olfactory projections which are conducted by
the medial bundle of the medial olfactory tract (Hamdani et al., 2000;
Døving and Lastein, 2009); in zebrafish, these projections integrate
into a circuit of amygdala-like structures in the telencephalon, as well
as hypothalamic structures (Gayoso et al., 2011, 2012). Serotonergic in-
nervation of such structures has been described in zebrafish (Kaslin and
Panula, 2001; Lillesaar et al., 2009), as well as monoamine oxidase
activity and immunoreactivity (Anichtchik et al., 2006), the presence
of 5-HT1A-like, 5-HT1B and 5-HT2C receptors (Norton et al., 2008;
Schneider et al., 2012), and serotonin transporters (Norton et al., 2008).
To better understand whether the serotonergic system participates in
the alarm response in zebrafish, we analyzed 5-HT content in the extracel-
lular fluid (ECF) of zebrafish exposed to conspecific skin extract, as well as
the effects of the serotonin transporter antagonist fluoxetine and the 5-
HT1AR antagonist WAY 100,635 on the behavioral responses of conspecific
skin extract in the light/dark assay and in a nocifensive behavior assay.
2. Methods
2.1. Drugs and reagents
Fluoxetine was bought from Roche (Brazil). WAY 100,635 maleate
was bought from Sigma (St. Louis, USA). HPLC standards (NE, Epi,
5-HT, 5-HIAA and DHBA) were bought from Sigma (St. Louis, USA).
HPLC-grade methanol was bought from Tedia (Fairfield, USA), and
biotechnology-grade sodium dodecyl sulfate was bought from Amresco
(Solon, USA). Reagents for hemoglobin quantification were bought from
LabTest (Brazil). All other reagents were of analytical-grade and bought
from Synth (Diadema, Brazil).
2.2. Animals
Wild-type longfin (n = 140) adult zebrafish were bought from a
local ornamental aquarium shop and collectively maintained in 40 L
tanks for at least two weeks before onset of experiments. The tanks
were kept at constant temperature (28 °C), oxygenation, and light
cycle (14:10 LD photoperiod). The animals were fed daily with Alcon
flake food, and twice a week with live brine shrimp.
2.3. Drug treatment
30 min before exposure to either conspecific skin extract or water, the
animals were cold-anesthetized and transferred to a sponge soaked in cold
water for intraperitoneal injection of 5 μl of either 2.5 mg/kg fluoxetine
hydrochloride, 0.003 mg/kg WAY 100635, or Cortland's salt solution (ve-
hicle). The animals were then transferred to a 2 L beaker to recover.
2.4. Behavioral assays
2.4.1. Alarm substance preparation and exposure
Alarm substance was produced from a conspecific skin extract as
described elsewhere (Speedie and Gerlai, 2008). Excess water was
removed from the skin of donor animals with a paper towel, after
which the animals were cold-anesthetized and quickly sacrificed by de-
capitation with surgical scissors. 10–15 shallow cuts were made on each
side of the trunk of 10 donor fish, after which the cuts were washed with
Milli-Q water. During the collection process and until further use, the
solution was kept on ice. A total of 100 ml of alarm substance solution
was collected, and the aliquots were diluted to a 50% concentration. Ex-
posure was performed in a pre-treatment beaker (3.5 ml skin extract/l)
for 5 min, after which the fish were removed from the exposure beaker
and subjected to one of the behavioral tests.
2.4.2. Scototaxis
The light/dark assay was performed as described elsewhere (Araujo
et al., 2012). Briefly, the animals were transferred to the central compart-
ment of a black and white tank (15 cm × 10 cm × 45 cm h × d × l) for a 3-
min. acclimation period, after which the doors which delimit this
compartment were removed and the animal was allowed to freely
explore the apparatus for 15 min. While the whole experimental tank
was illuminated from above by a homogeneous light source, due to
the reflectivity of the apparatus walls and floor average illumination
(measured just above the water line) above the black compartment
was 225 ± 64.2 (mean ± S.D.) lx, while in the white compartment it
was 307 ± 96.7 lx. The following variables were recorded:
time on the white compartment: the time spent in the white half of
the tank (percentage of the trial);
squares crossed: the number of 10 cm2 squares crossed by the animal
in the white compartment; latency to white: the time to first entry in
the white compartment (s);
entries in white compartment: the number of entries the animal
makes in the white compartment in the whole session;
erratic swimming: the number of “erratic swimming” events, defined
as a zig-zag, fast, unpredictable course of swimming of short
duration;
freezing: the proportional duration of freezing events (in % of time in
the white compartment), defined as complete cessation of move-
ments with the exception of eye and operculae movements;
thigmotaxis: the proportional duration of thigmotaxis events (in % of
time in the white compartment), defined as swimming in a distance
of 2 cm or less from the white compartment's walls;
risk assessment: the number of “risk assessment” events, defined as a
fast (b 1 s) entry in the white compartment followed by re-entry in
the black compartment, or as a partial entry in the white compart-
ment (i.e., the pectoral fin does not cross the midline);
Video records of the experiments were manually registered by two
observers blind to treatment (inter-observer reliability N 0.85) using
X-Plo-Rat 2005 (http://scotty.ffclrp.usp.br).
2.4.3. Fear-induced analgesia
The nocifensive behavior assay was modified from methods pro-
posed elsewhere (Correia et al., 2011; Maximino, 2011; Alves et al.,
2013). The locomotor activity of individual animals (n = 8–10) was re-
corded for 10 min before drug treatment to produce a baseline, after
which the animals were subjected to drug treatment and exposure to
alarm substance as above. After that, the animals were removed from
the exposure beaker and quickly injected with 5 ml of Cortland's salt
solution with 1% acetic acid in the tail in a region near the adipose fin.
Immediately after injection, the animals were transferred to the obser-
vation tank (a 10 cm l × 10 cm w × 20 cm h Plexiglas tank containing
water from the home tank) and their behavior was video recorded for
10 min. Digitized video files were then analyzed and two categories
were scored: tail-beating (the number of tail-beat movements that did
not lead to propulsion in the water) and locomotion (number of 9 cm3
squares crossed). These latter measures were normalized on a fish-by-
fish basis to the means of their individual pre-treatment values and
expressed as percentages, so all locomotion data in this assay refers to
changes relative to pre-treatment values.
 C. Maximino et al. / Pharmacology, Biochemistry and Behavior 124 (2014) 425–433
2.5. Physiological parameters
2.5.1. Sample preparation
After behavioral analyses, zebrafish were cold-euthanized and the
caudal fin and head were severed with a pair of fine scissors. Each fish
(n = 4 per group) was put into a homemade 0.5 ml microcentrifuge
tube that was previously perforated with sharp needles. This tube was
then placed into a 1.5 ml microcentrifuge tube containing 500 UI/ml
heparin. The assembly was then centrifuged at 50 g for 5 min at 11 °C.
After this step, another cut was made closely behind the existing
wound, removing the clot that had formed therein. Centrifugation was
repeated with the same parameters, yielding total blood at the bottom
tube (8.4 ± 1.1 μl) (Babaei et al., 2013).
While blood samples were prepared, another experimenter extract-
ed cerebrospinal fluid (CSF) from the severed heads. 0.5 ml of CSF was
extracted by quickly removing one brain from the skull and incubating
it in 2 ml of 50 mM TBS, pH 7.4, containing 90 mM NaCl, 2.5 mM CaCl2,
and 1 mM glutathione for 30 min at 4 °C (Pradel et al., 1999). This fluid
was then mixed with 0.5 ml of eluting solution, filtered through a
0.22 μm syringe filter and frozen at −20 °C until use.
2.5.2. Plasma glucose and hemoglobin levels
4 μl from the blood samples was thawed and used for glucose quanti-
fication, using Accu-Check Advantage II blood glucose strips and the Accu-
Check Advantage monitor (Roche Diagnostics, Germany). Another 4 μl
was added with 1 ml of Drabkin's solution (LabTest, Brazil) and measured
spectrophotometrically at 540 nm; hemoglobin levels were compared
against a cyanohemoglobin standard curve. All values were corrected by
protein levels, according to the method of Zor and Selinger (1996).
2.5.3. HPLC analysis of monoamines
Serotonin, 5-HIAA, norepinephrine, epinephrine and 3,4-
dihydroxybenzylamine (DHBA) (50 mg) were dissolved in 100 mL of
eluting solution (HClO4 70% [0.2 N], 10 mg EDTA, 9.5 mg sodium
metabissulfite) and frozen at -20 °C, to later be used as standards. The
HPLC system consisted of a delivery pump (LC20-AT, Shimadzu), a 20
μL sample injector (Rheodyne), a degasser (DGA-20A5), and an analytical
column (Shimadzu Shim-Pack VP-ODS, 250 x 4.6 mm internal diameter).
The integrating recorder was a Shimadzu CBM-20A (Shimadzu, Kyoto,
Japan). An electrochemical detector (Model L-ECD-6A) with glassy carbon
was used at a voltage setting of +0.72 V, with a sensitivity set at 2 nA full
deflection. The mobile phase consisted of a solution of 70 mM phosphate
buffer (pH 2.9), 0.2 mM EDTA, 34.6765 mM SDS, 10% HPLC-grade metha-
nol and 20% sodium metabissulfite as a conservative. The column temper-
ature was set at 17 °C, and the isocratic flow rate was 1.6 ml/min.
2.6. Statistical analysis
Categorical data (entries on white, squares crossed on white, erratic
swimming, risk assessment, tail-beating) were represented by boxplots
with Tukey's whiskers, and analyzed using Kruskal–Wallis ANOVAs.
Since latency estimates do not reach criteria for parametric analyses,
values were analyzed by the Mantel–Cox test and are represented as
Kaplan–Meier transformations in a survival plot (Jahn-Eimermacher
et al., 2011). The other data were represented as means ± standard
errors and analyzed by analyses of variance. p-Values b 0.05 were
considered statistically significant. In addition to p-values, replication
probabilities (Killeen, 2005) were also calculated for the contrasts,
using a built-in function from the R package ‘psych’ (v. 1.2.4).
3. Results
Alarm substance significantly reduced the time spent on the white
compartment of the light/dark test (Fig. 1A), an effect which was
avoided by pre-treatment with fluoxetine; fluoxetine itself decreased
this variable (F3, 39 = 5. 299, p = 0.004, prep = 0.914). Likewise,
427
alarm substance increased the latency to enter the white compartment
(Fig. 1B), an effect which once again was blocked by fluoxetine pre-
treatment (χ23 = 8.217, p = 0.0417, prep = 0.890). No effect was
observed in the number of entries in the white compartment (Fig. 1C;
H3 = 3.315, p = 0.3455, prep = 0.611) or number of squares crossed
in the white compartment (Fig. 1D; H3 = 2.343, p = 0.5043, prep =
0.497). Fluoxetine treatment and alarm substance increased freezing
(F3, 33 = 11, p b 0.0001, prep = 0.978), while pre-treatment with fluox-
etine abolished the alarm substance-induced freezing (Fig. 1E).
The same was observed in relation to erratic swimming events (Fig. 1F;
H3 = 15.68, p = 0.0013, prep = 0.983); while fluoxetine itself increased
thigmotaxis (Fig. 1G), alarm substance did not alter this variable (F3, 33 =
4.093, p = 0.0151, prep = 0.876). Finally, none of the treatments altered
risk assessment (H3 = 2.866, p = 0.4128, prep = 0.562).
In the nocifensive behavior test, alarm substance significantly de-
creased tail-beating caused by acetic acid injection at the 4–6 min
time block (F6, 108 = 5.164, p = 0.0069, prep = 0.917 for the interaction
term); pre-treatment with fluoxetine did not block this effect (Fig. 2A).
Likewise, alarm substance blocked the decrease in activity at the 2–4
min and 8–10 min time blocks caused by acetic acid (F8, 135 = 2.242,
p = 0.028, prep = 0.781 for the interaction term). Again, pre-
treatment with fluoxetine did not block this effect (Fig. 2B).
Treatment with WAY 100,635 increased the time spent on the white
compartment (Fig. 3A), but did not block the decrease in this variable
in animals exposed to the alarm substance (F3, 35 = 19.92, p b 0.0001,
prep = 0.996). Likewise, pre-treatment with WAY 100,635 did not
decrease the alarm substance-elicited increase in latency to white
(Fig. 3B; χ2 = 9.048, p = 0.0278, prep = 0.912). No effects were
observed on entries on white nor squares crossed in the white compart-
ment (Fig. 3C and D; H3 b 6.72, p N 0.082, prep b 0.837). This drug
did not alter freezing duration (Fig. 3E), and did not have any effect on
the increase in freezing elicited by alarm substance (F3, 35 = 8.662,
p = 0.0002, prep = 0.963). A similar pattern was observed in erratic
swimming (Fig. 3F; H3 = 20.71, p b 0.0001, prep N 0.996) and thigmo-
taxis (Fig. 3G; F3, 35 = 7.12, p = 0.0009, prep = 0.946). At last, WAY
100,635 decreased risk assessment (Fig. 3H), but no other effects were
observed in this variable (H3 = 14.12, p = 0.0027, prep = 0.975).
Pre-treatment with WAY 100,635 reduced the analgesic effect of
alarm substance exposure on tail-beating (Fig. 4A) at the 4–6 and 6–8
min time blocks (F8, 135 = 14.28, p b 0.001, prep = 0.993 for the interac-
tion term) and completely blocked it on the activity decrease at the 2–4
and 8–10 min time blocks (F8, 135 = 2.941, p = 0.0228, prep = 0.830;
Fig. 4B).
Alarm substance, but not fluoxetine, increased blood glucose, hemo-
globin, epinephrine and norepinephrine levels (Table 1). Pre-treatment
with fluoxetine partially blocked the effects of alarm substance on blood
glucose and hemoglobin levels, and totally blocked the increases in
circulating epinephrine and norepinephrine levels. WAY 100,635 did
not have effects by itself, nor did it block the effects of alarm substance.
Finally, treatment with fluoxetine increased extracellular 5-HT con-
tent and blocked the alarm substance-elicited increase (Fig. 5A; F3, 31 =
7.879, p = 0.0006, prep = 0.954). While fluoxetine did not produce an
effect on ECF 5-HIAA, it blocked the alarm substance-elicited increase
in this metabolite (Fig. 5B; F3, 31 = 5.219, p = 0.0055, prep = 0.909).
WAY 100,635 did not increase extracellular 5-HT nor 5-HIAA content
by itself, nor did it block the CAS-elicited increases (Fig. 5C and D;
F3, 31 N 2.9, p b 0.05, prep N 0.877).
4. Discussion
In the present work, it was demonstrated that alarm substance
increases scototaxis in adult zebrafish, also increasing the latency to
enter the white compartment, freezing and erratic swimming, but
not risk assessment or thigmotaxis. While most papers using the
light/dark test report only white avoidance (e.g., Gebauer et al., 2011;
Maximino et al., 2011; Norton et al., 2011; Steenbergen et al., 2011), it
428 C. Maximino et al. / Pharmacology, Biochemistry and Behavior 124 (2014) 425–433
 C. Maximino et al. / Pharmacology, Biochemistry and Behavior 124 (2014) 425–433
Fig. 2. Fluoxetine does not block the analgesic effect of CAS. (A) Number of tail-beating
events per time bin; (B) Normalized activity. **, p b 0.01 vs. control; *, p b 0.05 vs. control.
has recently been demonstrated that risk assessment and thigmotaxis
are also sensitive to serotonergic drugs (e.g., risk assessment and
thigmotaxis; Maximino et al., 2013a). Moreover, zebrafish from the leop-
ard phenotype, which show decreased tissue serotonin in the brain and
increased serotonin metabolism (Maximino et al., 2013b) also display in-
creased risk assessment, but not thigmotaxis. Finally, risk assessment and
thigmotaxis are also decreased by anxiolytic drugs with different molecu-
lar targets, while anxiogenic drugs increase these measures (Maximino
et al., in press). Overall, these results suggest that, while the light/dark
test seems to model conflict-induced anxiety (Maximino et al., in press),
the effects of fear-inducing stimuli such as acute alarm substance alter
specific behavioral measures which are not usually affected by manipula-
tions which increase or decrease anxiety.
As reported elsewhere (Maximino, 2011), alarm substance also pro-
duced analgesia. Moreover, extracellular 5-HT and 5-HIAA levels in the
brain were increased after exposure to Schreckstoff and blood glucose, he-
moglobin, epinephrine and norepinephrine levels were also increased;
these latter results suggest increased sympathetic activity, which can pre-
cede or parallel the increase in whole-body cortisol that is observed in
these animals after CAS exposure (Schirmer et al., 2013). Fluoxetine treat-
ment blocked the effects of alarm substance on the light/dark assay and
Fig. 1. Fluoxetine, at an anxiogenic dose, blocks the behavioral effects of alarm substance on the light/dark test. (A) Time spent on the white compartment; (B) Latency to enter the
white compartment; (C) Number of entries in the white compartment; (D) Number of squares crossed in the white compartment; (E) Total duration of freezing events in the white
compartment; (F) Frequency of erratic swimming events in the white compartment; (G) Proportion of time on white spent in thigmotaxis; (H) Frequency of risk assessment events.
Black bars represent animals exposed to water, and gray bars animals exposed to conspecific alarm substance (CAS). Different letters indicate statistically significant differences.
429
on blood parameters, but did not block the fear-induced analgesia; con-
versely, WAY 100,635 did not block the effects of CAS on scototaxis or
blood parameters, but blocked fear-induced analgesia. Pre-treatment
with fluoxetine also blocked the alarm substance-elicited increase in 5-
HT and 5-HIAA, an effect which was not observed with WAY 100,635.
The role of serotonin transporter (SERT) on the behavioral responses
to fear stimuli has been investigated in rodents (Adamec et al., 2006;
Wellman et al., 2007; Ravinder et al., 2011), humans (Perkins et al.,
2009; Schruers et al., 2011) and fish (Beulig and Fowler, 2008;
Maximino et al., 2011; Roussigné et al., 2011; Barbosa et al., 2012). In
piauçu fish, pre-treatment with fluoxetine (10 mg/kg) decreases the
hypoactivity induced by exposure to conspecific skin extract (Barbosa
et al., 2012); the opposite effect was observed in Arabian killifish
(Barry, 2013). In mice, acute fluoxetine (5 mg/kg, but not higher
doses) increases escape responses in the Mouse Defense Test Battery
(Blanchard et al., 1997). Interestingly, while fluoxetine by itself
increased white avoidance in the scototaxis test in the present experi-
ments, it reversed the increase caused by alarm substance. The same
dose of fluoxetine was shown to decrease the exaggerated scototaxis
and risk assessment seen in leopard zebrafish in the light/dark test
(Maximino et al., 2013b). These results are consonant with the hypoth-
esis that, in zebrafish as well as in mammals, 5-HT increases anxiety and
decreases fear (Graeff et al., 1996, 1997; Zangrossi et al., 2001; Hale and
Lowry, 2011; Maximino et al., in press), but suggest that the serotoner-
gic modulation of zebrafish defensive behavior shows complexities
which are not explainable solely by this ‘dual role’ hypothesis.
The dose of fluoxetine used in the present experiment was enough
to increase extracellular 5-HT levels, but the increase induced by
alarm substance was quantitatively equivalent to that caused by
blocking SERT, which could indicate that this effect was mediated by
an action on SERT to increase 5-HT levels. Moreover, SERT inhibition
also blocked the effects of alarm substance on both 5-HT levels and
scototaxis, which could indicate that this effect was mediated by an ac-
tion on SERT. Whether this results from an inhibition of 5-HT transport
or a reversal of uptake is yet unclear, and further studies are necessary
to test this hypothesis.
While fluoxetine treatment blocked the effects of conspecific alarm
substance on scototaxis, it was not able to block the antinociceptive
effect of the alarm reaction. Analgesia during and after presentation of
predators and partial predator stimuli has been described in rodents
(Kavaliers and Douglas, 1991; Wiedenmayer and Barr, 1995) and fish
(Maximino, 2011; Lima et al., 2012; Alves et al., 2013). In rodents, opioid
and non-opioid components of fear-induced analgesia have been de-
scribed, with the non-opioid component being mediated by GABAergic
and serotonergic mechanisms (Kavaliers and Douglas, 1991). While an
opioid component has been described in the piauçu fish (Alves et al.,
2013), the role of serotonin has not yet been described. While fluoxetine
did not block the antinociceptive effect of alarm substance, pre-
treatment with the 5-HT1AR antagonist WAY 100,635 had this effect.
While WAY 100,635 decreased scototaxis and risk assessment by itself,
it had no effect on the alarm substance-induced increase in white avoid-
ance, latency to white, freezing, or erratic swimming. While 5-HT1A partial
agonists and full antagonists have been described as decreasing anxiety-
like behavior in different tests and species (Zangrossi et al., 2001;
Herculano and Maximino, in press), the role of these receptors in fear re-
sponses is less certain. In zebrafish, scototaxis and geotaxis are equally de-
creased by WAY 100,635 and buspirone (Maximino et al., 2013a). 5-HT1A
receptor knockout mice show normal conditional fear (Groenink et al.,
2003), and 5-HT1A partial agonists do not affect one-way escape in the
430 C. Maximino et al. / Pharmacology, Biochemistry and Behavior 124 (2014) 425–433
 C. Maximino et al. / Pharmacology, Biochemistry and Behavior 124 (2014) 425–433 431

From a pharmacological point of view, 5-HT1A partial agonists ef-

fectively act as full antagonists at postsynaptic sites and full agonists
at presynaptic sites. In rats, microinjection of 8-OH-DPAT in the
amygdala does not block stress-induced analgesia (Nunes-de-
Souza et al., 2000), but decreases one-way escape in the elevated
T-maze (Guimarães et al., 2010); reduction of one-way escape is
also observed with microinjection in the periaqueductal gray, while
microinjection of 8-OH-DPAT in the hippocampus and lateral sep-
tum increases inhibitory avoidance in the elevated T maze
(Guimarães et al., 2010). Since 5-HT levels were not altered by
WAY 100,635 in the present experiments, it is suggested that post-
synaptic 5-HT1A receptors mediate fear-induced analgesia, but not
other behavioral effects of alarm substance.
Fluoxetine treatment also blocked the sympathetic burst elicited
by CAS, while WAY 100,635 had no such effect. These results sug-
gest that the fear-like reaction to CAS is mediated by a serotonin-
induced sympathetic activation, and that this reaction is not
mediated by the 5HT1A receptor. These results are consistent with
the observation that treatment with WAY 100,635 reduces condi-
tioned ultrasonic vocalizations in rats without altering plasmatic
ACTH, corticosterone, prolactin or glucose levels (Groenink et al.,
1996). They are also consistent with the observation that footshock-
induced analgesia is independent on sympathetic nervous system
activity (Watkins et al., 1982).
Overall, these results suggest that, in zebrafish, fear-induced in-
creases in defensive behavior and sympathetic activation are mediated
by the serotonin transporter, while fear-induced analgesia is mediated
by the 5-HT1A receptor. The precise site of action of these drugs,
however, awaits further experiments. The present work underlies the
evolutionary conservation of the role of serotonin in fear-like behavior
in vertebrates, opening avenues for future investigations in the genetic
underpinnings of these functions.
Fig. 4. WAY 100,635 blocks the analgesic effect of CAS. (A) Number of tail-beating events
per time bin; (B) Normalized activity. ***, p b 0.001 vs. control; **, p b 0.01 vs. control;
*, p b 0.05 vs. control.
Acknowledgments

This work was partially supported by the CNPq (Brazil) (Grant

number 483336/2009-2). AMH is a recipient of a CAPES produc-
rat elevated T-maze (Zangrossi et al., 2001). Nonetheless, the 5-HT1A full tivity grant. CM and MGL were recipients of a CAPES scholarship
agonist 8-OH-DPAT decreases the threshold for periaqueductal gray when the work was executed, and CCC was a recipient of a PROEX/
stimulation-induced escape responses (Jenck et al., 1989). UFPA grant.

Table 1
Fluoxetine, but not WAY 100,635, blocks sympathetic activation elicited by CAS, as measured by blood catecholamine (epinephrine and norepinephrine), hemoglobin and glucose levels.

Parameter Control CAS Drug CAS + drug Statistical analysis

Fluoxetine
Blood glucose (mg/dl) 50.3 ± 16.25 121.8 ± 48.6⁎⁎ 48.1 ± 16.09 75.3 ± 19.21⁎ F3, 15 = 9.052, p = 0.0021
Hemoglobin (mmol/l) 4.4 ± 1.4 8.3 ± 0.9⁎⁎ 4.32 ± 0.3 6.7 ± 0.7⁎ F3, 15 = 29.14, p b 0.0001
Norepinephrine (pg/ml) 803.4 ± 82.7 1245.7 ± 128.3⁎⁎ 851.5 ± 87.7 769.7 ± 79.3 F3, 15 = 11.83, p = 0.0007
Epinephrine (pg/ml) 400.1 ± 120.1 1015.1 ± 123.7⁎⁎⁎ 357.4 ± 103.1 451.1 ± 58.1 F3, 15 = 29.53, p b 0.0001

WAY 100,635
Blood glucose (mg/dl) 51.3 ± 3.1 132.1 ± 23.1⁎⁎ 57.5 ± 8.3 142.1 ± 31.2⁎⁎ F3, 15 = 17.54, p = 0.0001
Hemoglobin (mmol/l) 4.8 ± 0.3 9.1 ± 1.3⁎⁎⁎ 3.9 ± 1.5 10.2 ± 0.8⁎⁎⁎ F3, 15 = 29.9, p b 0.0001
Norepinephrine (pg/ml) 798.7 ± 81.1 1232.1 ± 118.3⁎⁎⁎ 815.1 ± 58.9 1313.6 ± 129.4⁎⁎⁎ F3, 15 = 54.4, p b 0.0001
Epinephrine (pg/ml) 403.5 ± 98.7 1021.9 ± 98.1⁎⁎⁎ 420.1 ± 101.7 997.1 ± 81.4⁎⁎⁎ F3, 15 = 42.24, p b 0.0001
⁎⁎⁎ p b 0.001 vs. control.
⁎⁎ p b 0.01 vs. control.
⁎ p b 0.05 vs. control.

Fig. 3. WAY 100,635, at an anxiolytic dose, does not block the behavioral effects of CAS on the light/dark test. (A) Time spent on the white compartment; (B) Latency to enter the white
compartment; (C) Number of entries in the white compartment; (D) Number of squares crossed in the white compartment; (E) Total duration of freezing events in the white compartment;
(F) Frequency of erratic swimming events in the white compartment; (G) Proportion of time on white spent in thigmotaxis; (H) Frequency of risk assessment events. Black bars represent
animals exposed to water, and gray bars animals exposed to conspecific alarm substance (CAS). Different letters indicate statistically significant differences.
432 C. Maximino et al. / Pharmacology, Biochemistry and Behavior 124 (2014) 425–433
Fig. 5. Fluoxetine, but not WAY 100,635, blocks the CAS-elicited increases in extracellular serotonin in the brain. Extracellular 5-HT levels in water- (black bars) and CAS-exposed (gray
bars) animals treated with (A) fluoxetine or (C) WAY 100,635; extracellular 5-HIAA levels in water- (black bars) and CAS-exposed (gray bars) animals treated with (B) fluoxetine or
(D) WAY 100,635.
References
Adamec R, Burton P, Blundell J, Murphy DL, Holmes A. Vulnerability to mild predator
stress in serotonin transporter knockout mice. Behav Brain Res 2006;170:126–40.
Alves FL, Barbosa Jr A, Hoffmann A. Antinociception in piauçu fish induced by exposure to
the conspecific alarm substance. Physiol Behav 2013;110–111:58–62.
Anichtchik O, Sallinen V, Peitsaro N, Panula P. Distinct structure and activity of monoamine
oxidase in the brain of zebrafish (Danio rerio). J Comp Neurol 2006;610:593–610.
Araujo J, Maximino C, de Brito TM, da Silva AWB, Batista Ede JO, Oliveira KRM, et al.
Behavioral and pharmacological aspects of anxiety in the light/dark preference test.
In: Kalueff AV, Stewart AM, editors. Zebrafish protocols for neurobehavioral research.
New York: Humana Press; 2012.
Babaei F, Ramalingam R, Tavendale A, Liang Y, Yan LSK, Ajuh P, et al. Novel blood collec-
tion method allows plasma proteome analysis from single zebrafish. J Proteome Res
2013;12:1580–90.
Barbosa Jr A, Alves FL, Pereira Ade SF, Ide LM, Hoffmann A. Behavioral characterization of
the alarm reaction and anxiolytic-like effect of acute treatment with fluoxetine in
piauçu fish. Physiol Behav 2012;105:784–90.
Barry MJ. Effects of fluoxetine on the swimming and behavioural responses of the Arabian
killifish. Ecotoxicology 2013;22(2):425–32.
Beulig A, Fowler J. Fish on Prozac: effect of serotonin reuptake inhibitors on cognition in
goldfish. Behav Neurosci 2008;122:426–32.
Blanchard RJ, Griebel G, Henrie JA, Blanchard DC. Differentiation of anxiolytic and
panicolytic drugs by effects on rat and mouse defense test batteries. Neurosci
Biobehav Rev 1997;21:783–9.
Blaser RE, Goldsteinholm K. Depth preference in zebrafish, Danio rerio: control by surface
and substrate cues. Anim Behav 2012;83:953–9.
Blaser RE, Peñalosa YM. Stimuli affecting zebrafish (Danio rerio) behavior in the light/dark
preference test. Physiol Behav, 104. Elsevier Inc.; 2011. p. 831–7.
Blaser RE, Chadwick L, McGinnis GC. Behavioral measures of anxiety in zebrafish
(Danio rerio). Behav Brain Res 2010;208:56–62.
Correia AD, Cunha SR, Scholze M, Stevens ED. A novel behavioral fish model of
nociception for testing analgesics. Pharmaceuticals 2011;4:665–80.
Deakin JFW, Graeff FG. 5-HT and mechanisms of defense. J Psychopharmacol 1991;5:305–15.
Døving KB, Lastein S. The alarm reaction in fishes — odorants, modulation of responses,
neural pathways. Ann N Y Acad Sci 2009;1170:413–23.
Egan RJ, Bergner CL, Hart PC, Cachat JM, Canavello PR, Elegante MF, et al. Understanding
behavioral and physiological phenotypes of stress and anxiety in zebrafish. Behav
Brain Res 2009;205:38–44.
Gayoso JÁ, Castro A, Anadón R, Manso MJ. Differential bulbar and extrabulbar projections
of diverse olfactory receptor neuron populations in the adult zebrafish (Danio rerio).
J Comp Neurol 2011;519:247–76.
Gayoso JÁ, Castro A, Anadón R, Manso MJ. Crypt cells of the zebrafish Danio rerio mainly
project to the dorsomedial glomerular field of the olfactory bulb. Chem Senses 2012;
37:357–69.
Gebauer DL, Pagnussat N, Piato ÂL, Schaefer IC, Bonan CD, Lara DR. Effects of anxiolytics in
zebrafish: Similarities and differences between benzodiazepines, buspirone and eth-
anol. Pharmacol Biochem Behav 2011;99:480–6.
Gerlai R. Zebrafish antipredatory responses: a future for translational research? Behav
Brain Res 2010;207:223–31.
Graeff FG, Guimarães FS, De Andrade TGCS, Deakin JFW. Role of 5-HT in stress, anxiety,
and depression. Pharmacol Biochem Behav 1996;54:129–41.
Graeff FG, de Viana MB, Mora PO. Dual role of 5-HT in defense and anxiety. Neurosci
Biobehav Rev 1997;21:791–9.
Gray JA, McNaughton N. The neuropsychology of anxiety: an enquiry into the functions of
the septo-hippocampal system. 2nd ed. Oxford: Oxford University Press; 2000.
Groenink L, Mos J, van der Gugten J, Olivier B. The 5-HT1A receptor is not involved in
emotional stress-induced rises in stress hormones. Pharmacol Biochem Behav
1996;55:303–8.
Groenink L, Pattij T, De Jongh R, Van der Gugten J, Oosting RS, Dirks A, et al. 5-HT1A recep-
tor knockout mice and mice overexpressing corticotropin-releasing hormone in
models of anxiety. Eur J Pharmacol 2003;463:185–97.
Guimarães FS, Zangrossi Jr H, Del-Ben CM, Graeff FG. Serotonin in panic and anxiety
disorders. In: Müller C, Jacobs B, editors. Handb behav neurobiol serotonin.
Amsterdam: Elsevier B. V.; 2010. p. 667–85.
Hale MW, Lowry CA. Functional topography of midbrain and pontine serotonergic systems:
implications for synaptic regulation of serotonergic circuits. Psychopharmacology 2011;
213:243–64.
 C. Maximino et al. / Pharmacology, Biochemistry and Behavior 124 (2014) 425–433
Hamdani E, Stabell OB, Alexander G, Døving KB. Alarm reaction in the crucian carp is
mediated by the medial bundle of the medial olfactory tract. Chem Senses 2000;25:
103–9.
Herculano AM, Maximino C. Serotonergic modulation of zebrafish behavior: towards a
paradox. Prog Neuropsychopharmacol Biol Psychiatry 2014. (in press).
Jahn-Eimermacher A, Lasarzik I, Raber J. Statistical analysis outcomes in behavioral exper-
iments. Behav Brain Res 2011;221:271–5.
Jenck F, Broekkamp CL, Van Delft AM. Opposite control mediated by central 5-HT1A
and non-5HT1A (5-HT1B or 5-HT1C) receptors on periaqueductal gray aversion.
Eur J Pharmacol 1989;161:219–21.
Jesuthasan SJ, Mathuru AS. The alarm response in zebrafish: innate fear in a vertebrate
genetic model. J Neurogenet 2008;22:211–29.
Kalueff AV, Stewart AM, Kyzar EJ, Cachat J, Gebhardt M, Landsman S, et al. Time to
recognize zebrafish “affective” behavior. Behaviour 2012;149:1019–36.
Kaslin J, Panula P. Comparative anatomy of the histaminergic and other aminergic
systems in zebrafish (Danio rerio). J Comp Neurol 2001;440:342–77.
Kavaliers M, Douglas DC. Sex differences in opioid and non-opioid mediated predator-
induced analgesia in mice. Brain Res 1991;568:173–7.
Killeen PR. An alternative to null-hypothesis significance tests. Psychol Sci 2005;16:
345–53.
Lillesaar C, Stigloher C, Tannhäuser B, Wullimann MF, Bally-Cuif L. Axonal projections
originating from raphe serotonergic neurons in the developing and adult zebrafish,
Danio rerio, using transgenics to visualize raphe-specific pet1 expression. J Comp
Neurol 2009;512:158–82.
Lima MG, Maximino C, Batista Ede JO, Oliveira KRM, Herculano AM. Nocifensive behavior
in adult and larval zebrafish. In: Kalueff AV, Stewart AM, editors. Zebrafish protocols
for neurobehavioral research. New York: Humana Press; 2012.
Luca RM, Gerlai R. In search of optimal fear inducing stimuli: differential behavioral
responses to computer animated images in zebrafish. Behav Brain Res 2012;226:
66–76.
Mathuru AS, Kibat C, Cheong WF, Shui G, Wenk MR, Friedrich RW, et al. Chondroitin
fragments are odorants that trigger fear behavior in fish. Curr Biol 2012;22:
538–44.
Maximino C. Modulation of nociceptive-like behavior in zebrafish (Danio rerio) by
environmental stressors. Psychol Neurosci 2011;4:149–55.
Maximino C, de Brito TM, Colmanetti R, Pontes AAA, de Castro HM, de Lacerda RIT,
et al. Parametric analyses of anxiety in zebrafish scototaxis. Behav Brain Res
2010;210:1–7.
Maximino C, de Silva AWB, Gouveia Jr A, Herculano AM. Pharmacological analysis of
zebrafish (Danio rerio) scototaxis. Prog Neuro-Psychopharmacol Biol Psychiatry
2011;35(2):624–31.
Maximino C, Puty B, Benzecry R, Araújo J, Lima MG, Batista EJO, et al. Role of serotonin in
zebrafish (Danio rerio) anxiety: relationship with serotonin levels and effect of
buspirone, WAY 100635, SB 224289, fluoxetine and para-chlorophenylalanine
(pCPA) in two behavioral models. Neuropharmacology 2013a;71:83–97.
Maximino C, Puty B, Oliveira KRM, Herculano AM. Behavioral and neurochemical changes
in the zebrafish leopard strain. Genes Brain Behav 2013b;12:576–82.
Maximino C, da Silva AWB, Araújo J, Lima MG, Miranda V, Puty B, et al. Fingerprinting of
psychoactive drugs in zebrafish anxiety-like behaviors. PLoS ONE 2014. (in press).
McNaughton N, Corr PJ. A two-dimensional neuropsychology of defense: fear/anxiety and
defensive distance. Neurosci Biobehav Rev 2004;28:285–305.
433
Norton WHJ, Folchert A, Bally-Cuif L. Comparative analysis of serotonin receptor (HTR1A/
HTR1B families) and transporter (slc6a4a/b) gene expression in the zebrafish brain.
J Comp Neurol 2008;511:521–42.
Norton WHJ, Stumpenhorst K, Faus-Kessler T, Folchert A, Rohner N, Harris MP, et al.
Modulation of Fgfr1a signaling in zebrafish reveals a genetic basis for the aggression-
boldness syndrome. J Neurosci 2011;31:13796–807.
Nunes-de-Souza RL, Canto-de-Souza A, Da-Costa M, Fornari RV, Graeff FG, Pelá IR.
Anxiety-induced antinociception in mice: effects of systemic and intra-amygdala ad-
ministration of 8-OH-DPAT and midazolam. Psychopharmacology 2000;150:300–10.
Perkins AM, Ettinger U, Davis R, Foster R, Williams SCR, Corr PJ. Effects of lorazepam and
citalopram on human defensive reactions: ethopharmacological differentiation of
fear and anxiety. J Neurosci 2009;29:12617–24.
Pradel G, Schachner M, Schmidt R. Inhibition of memory consolidation by antibodies
against cell adhesion molecules after active avoidance conditioning in zebrafish.
J Neurobiol 1999;39:197–206.
Ravinder S, Pillai AG, Chattarji S. Cellular correlates of enhanced anxiety caused by acute
treatment with the selective serotonin reuptake inhibitor fluoxetine in rats. Front
Behav Neurosci 2011;5. (Article 88).
Roussigné M, Blader P, Wilson SW. Breaking symmetry: the zebrafish as a model for
understanding left–right asymmetry in the developing brain. Dev Neurobiol 2011;
72:269–81.
Sackerman J, Donegan JJ, Cunningham CS, Nguyen NN, Lawless K, Long A, et al. Zebrafish
behavior in novel environments: effects of acute exposure to anxiolytic compounds
and choice of Danio rerio line. Int J Comp Psychol 2010;23(1):43–61.
Schirmer A, Jesuthasan S, Mathuru AS. Tactile stimulation reduces fear in fish. Front Behav
Neurosci 2013;7. (Article 167).
Schneider H, Fritzky L, Williams J, Heumann C, Yochum M, Pattar K, et al. Cloning and
expression of a zebrafish 5-HT2C receptor gene. Gene 2012;502:108–17.
Schruers K, Esquivel G, Van Duinen M, Wichers M, Kenis G, Colasanti A, et al. Genetic
moderation of CO2-induced fear by 5-HTTLPR genotype. J Psychopharmacol 2011;
25:37–42.
Speedie N, Gerlai R. Alarm substance induced behavioral responses in zebrafish
(Danio rerio). Behav Brain Res 2008;188:168–77.
Steenbergen PJ, Richardson MK, Champagne DL. Patterns of avoidance behaviours in the
light/dark preference test in young juvenile zebrafish: a pharmacological study.
Behav Brain Res 2011;222:15–25.
Stewart A, Wu N, Cachat J, Hart P, Gaikwad S, Wong K, et al. Pharmacological modu-
lation of anxiety-like phenotypes in adult zebrafish behavioral models. Prog
Neuropsychopharmacol Biol Psychiatry 2011;35:1421–31.
Watkins LR, Cobelli DA, Newsome HH, Mayer DJ. Footshock induced analgesia is depen-
dent neither on pituitary nor sympathetic activation. Brain Res 1982;245:81–96.
Wellman CL, Izquierdo A, Garrett JE, Martin KP, Carroll J, Millstein R, et al. Impaired stress-
coping and fear extinction and abnormal corticolimbic morphology in serotonin
transporter knock-out mice. J Neurosci 2007;27:684–91.
Wiedenmayer CP, Barr GA. Developmental changes in responsivity to threat are stimulus-
specific in rats. Dev Psychobiol 1995;39:1–7.
Zangrossi Jr H, de Viana MB, Zanoveli J, Bueno C, Nogueira RL, Graeff FG. Serotonergic
regulation of inhibitory avoidance and one-way escape in the rat elevated T-maze.
Neurosci Biobehav Rev 2001;25:637–45.
Zor T, Selinger Z. Linearization of the Bradford protein assay increases its sensitivity:
theoretical and experimental studies. Anal Biochem 1996;236:302–8.